Page 10 of 17

Re: Experiments..

Posted: Sat Oct 09, 2010 7:37 pm
by neo0.
Atmospheric scattering! Yess!

Will it be possible to replicate things such as a sun set?

Re: Experiments..

Posted: Sat Oct 09, 2010 8:16 pm
by lycium
sure, look at the pictures 8)

Re: Experiments..

Posted: Sun Oct 10, 2010 12:26 am
by OnoSendai
Zom-B wrote:looks great! Reminds me of http://nis-lab.is.s.u-tokyo.ac.jp/~egak ... sia10.html

2 questions:

1) Indigo 2.4.11 ??
2) whats the banding on the snow in the canyon... shouldn't it be smooth through 32bit depth?
1. It's a trunk build (main dev version).. not really 2.4.11, just haven't updated the internal version number in trunk yet.

2. Not sure what's up with that. the displacement values should at least be interpolated.

Re: Experiments..

Posted: Sun Oct 10, 2010 12:29 am
by OnoSendai
Zom-B wrote:rayleight for correct dawn sky color and mie scattering for a atmospheric effect, this enhancements are very welcome, since the actual sky system is quite... unspectacular... to use ;)

About the mie scattering: is this a global value for the sky, where to define the size of (water!?) particle in the air to get a denser misty look, or is this a material based feature?
How can a user control this... is a giant sphere of earth atmosphere always needed? How does this perform with BiDir, since it gets less performance if scenes are BIIIIG afaik!?
The amount of Mie scattering is affected by the turbidity of the atmosphere, which can be set by the user in the same was as the current sky model.
For now we are using giant Earth-spheres, but we could probably get away with just a patch of the Earth's surface :)
We only use path tracing for these experiments, as bidir can't cope :)

Re: Experiments..

Posted: Mon Oct 11, 2010 1:02 pm
by StompinTom
That sunset looks amazing! Any way of generalizing that into a new sky model? Sunsets just don't look as great with the current model.

Re: Experiments..

Posted: Mon Oct 11, 2010 3:11 pm
by neo0.
StompinTom wrote:That sunset looks amazing! Any way of generalizing that into a new sky model? Sunsets just don't look as great with the current model.
Agreed.

Re: Experiments..

Posted: Wed Oct 13, 2010 9:52 pm
by OnoSendai
Made a little news post about atmospheric scattering:

http://www.indigorenderer.com/node/1143

Re: Experiments..

Posted: Wed Oct 27, 2010 7:01 pm
by CoolColJ
Unbiased Terragen :)

Re: Experiments..

Posted: Fri Oct 29, 2010 12:07 am
by hcpiter
OnoSendai wrote:Made a little news post about atmospheric scattering:

http://www.indigorenderer.com/node/1143
awesome environment !
very well ! 2nd render is good for desktop background with 2 screens.

Re: Experiments..

Posted: Fri Oct 29, 2010 12:29 am
by lycium
hcpiter wrote:2nd render is good for desktop background with 2 screens.
exactly how it's intended :)

Re: Experiments..

Posted: Sat Feb 26, 2011 6:51 am
by OnoSendai
A new test render :)

Re: Experiments..

Posted: Sat Feb 26, 2011 6:53 am
by OnoSendai
pre-dawn render:

Re: Experiments..

Posted: Sat Feb 26, 2011 7:19 am
by StompinTom
HOW many samples/sec is that? :D The pre-dawn looks awesome! Improved sky model? Is it any more flexible too? Meaning, are there more parameters to tweak the color, haze and stuff?

Re: Experiments..

Posted: Sat Feb 26, 2011 7:35 am
by OnoSendai
The only tweakable parameter currently is turbidity, which controls the amount of water vapour and hence Mie scattering.

Re: Experiments..

Posted: Sat Feb 26, 2011 7:42 am
by StompinTom
Other engines have a Ozone setting in their model too, but I never really figure out what it does... I'm assuming the Indigo sky model doesn't use such a thing?